EGM‐Guided Ablation: CFE, Ganglia, Rotors… ISHNE 2009 At lV MD FRCPC AtulVerma, MD FRCPC Staff Cardiologist, Electrophysiologist Southlake Regional Health Centre k d Newmarket, Canada
EGM‐Guided Ablation:CFE, Ganglia, Rotors…ISHNE 2009
At l V MD FRCPCAtul Verma, MD FRCPCStaff Cardiologist, Electrophysiologist
Southlake Regional Health Centrek dNewmarket, Canada
DisclosuresDisclosures
Consulting fees/honorariag /– Biosense Webster, Medtronic, St Jude Medical, Sanofi
Aventis, Astra Zeneca, Servier
S k bSpeakers bureau– Biosense Webster, Medtronic, St Jude Medical
Research grantsResearch grants– Biosense Webster, Medtronic, St Jude Medical
Equity, Ownership, Salary, Royaltiesq y, p, y, y– None
EGM‐Guided AblationEGM Guided Ablation
Complex Fractionated Electrograms (CFE)Complex Fractionated Electrograms (CFE)STAR AF dataGanglionated Plexi (GP)Ganglionated Plexi (GP)Dominant Frequency (DF)
Question #1Question #1
Of the following EGM‐based targets, which do you Of the following EGM based targets, which do you feel offer the greatest potential value as a hybrid strategy to PVI?
1. CFE2. Ganglionated Plexi3. Dominant Frequency3 q y4. None of the above
Question #2Question #2
In which AF patients do you think CFE are most In which AF patients do you think CFE are most useful to study/ablate?
1. All AF2. Paroxysmal AF2. Paroxysmal AF3. Persistent AF
Question #3Question #3
What do you think are the biggest barriers to using What do you think are the biggest barriers to using CFE ablation for AF (either alone or in combination)?
1. Not a consistent definition of CFEN t t bl t t f bl ti2. Not a stable target for ablation
3. Not enough data showing its benefith d l4. Too much extra procedural time
5. None of the above
Complex Fractionated Complex Fractionated Electrograms (CFE)
Theory Behind CFEsTheory Behind CFEs
• Intraoperative mapping of p pp gRA during induced AF• Identified 4 types of atrial potentials during AFpotentials during AF• Fragmented potentials correlated to pivot points
h lti l l twhere multiple wavelets arc around a region of functional block
Konings et al, Circulation 1997• OR areas of slow conduction
Definition of CFE – Nademanee JACC 2004Definition of CFE Nademanee JACC 2004
(A)atrial electrograms that are F ti t d & d f tFractionated & composed of two deflections or more andhave a perturbation of the baseline with continuous deflectionswith continuous deflections
(B) atrial electrograms with a(B) atrial electrograms with avery short cycle length (<120 ms) with or without multiple potentials when compared with the atrial CLpfrom other parts of the atria
Where are the CFE?Where are the CFE?
PVsR fRoof
Septump
RA
CS
Nademanee, JACC 2004Nademanee, JACC 2004
AFCL increased from:– 172 ± 26 ms to 237 ± 42 ms
AF termination during ablation:l d b l d– Paroxysmal – 100% (14% required ibutilide)
– Chronic – 91% (28% required ibutilide)
76% success rate after 1 procedure15% became arrhythmia‐free after a second procedureAt least 5% of these successful patients remained on AAD
Oral et al, Circulation 2007Oral et al, Circulation 2007
Prospective, non‐randomized assessmentProspective, non randomized assessmentCFE ablation only – no PV isolationN=100N=100Permanent AF patients only
Oral et al, Circulation 2007Oral et al, Circulation 2007
Very low acute termination rate of AF during ablationVery low acute termination rate of AF during ablation– Only 16% terminated acutely
Long‐term success rate of this approach modest– 57% AF‐free off medications– 44% required two procedures– Average of 14 months follow‐up
Reproducibility of CFE ResultsReproducibility of CFE Results
Results from other centers not consistentResults from other centers not consistentPart of the problem stems from definition and understanding of CFEgAre these targets really spatially and temporally stable?Is there a consistent way to define CFE?
Haissaguerre et al, H Rh h 6Heart Rhythm 2006
“CFE” are dependent on AFCLCFE are dependent on AFCLThe faster the CL, the more likely it is to see “CFE”Since AFCL can be variable so to can occurrence of Since AFCL can be variable, so to can occurrence of “CFE”Therefore, “elusive” “transient” targetTherefore, elusive transient target
Haissaguerre et al, H Rh h 6Heart Rhythm 2006
Problem: Are these “CFE”?
Automated CFE MappingAutomated CFE Mapping
Subjectivity of CFE definition still a major limitationSubjectivity of CFE definition still a major limitationAutomated algorithms are novel tools to standardize CFE definition and mappingpp gAverages the signal analysis over 4‐8 sec, allowing differentiation from transient vs stable CFE sites
Implementation and Deployment f h CFE M i T l (NAVX)of the CFE Mapping Tool (NAVX)Activation Detection Criteria
– Peak‐to‐peak voltage threshold• > Baseline noise floor
Sl th h ld
~.05 mV
– Slope threshold• Near field vs. far field
– Refractory settingA id d bl ti
~20 ms
• Avoid double counting
Map Display Representation
~50 ms
Organized
– Average interval representedwith color scale (ms)• 1‐8 second evaluation length
Organized
Fractionated
Diagnostic Landmarking: NavX System3D mapping from 3 pp gstandard cathetersPoints are collected at electrodes and projected onto map surfaceUsing MultiPoint technology, points may be collected from
A i l l t d– A single electrode– A multipolar catheter– All catheters in use
MultiPoint DxL mapping of fractionationImage courtesy of Dr Andrea Natale, CCF
Mapping CFE RegionsMapping CFE RegionsBy annotating the
deflections of the local EGM, a mean CL can be
calculated – CFE defined as regions with
CL < 120 msCL < 120 ms.
Courtesy of Dr. Koonlawee Nademanee, Los Angeles, CA
DxL Map Settings CFE MeanDxL Map Settings – CFE Mean
Recommended Settings
Width 10‐20 ms (avoid far‐field)
Refractory 30‐50 ms (avoid double counting)Refractory 30‐50 ms (avoid double counting)
P‐P Sensitivity 0.03‐0.05 mV (avoid noise)
Segment Length 4‐8 sec
Interpolation 4‐10 mm
Internal/External Projection 3‐6 mm
CFE Stability – Verma et al, H Rh h 8Heart Rhythm 2008
CFE maps taken at 0CFE maps taken at 0 and 20 minutes. Degree of overlap and
i t f CFEconsistency of CFE areas studied.
CFE definition – regions with local EGM cycle length <120 mslength <120 ms.
CFE Stability – Verma et al, H t Rh th 8Heart Rhythm 2008
80%
100%
80%
100%
0%
20%
40%
60%
0 min 20 min
CBA
0%
20%
40%
60%
0 min 20 min
C
B
A When measured at 0 and 20 minutes, proportion of signals within cycle
40%
60%
80%
100%
CBA 40%
60%
80%
100%
CBA
LPV RPV of signals within cycle length ranges are
stable.
0%
20%
40%
0 min 20 min
A
0%
20%
40%
0 min 20 min
A
SEPT PWBlack = 50-120 ms
Whit 121 200
20%
40%
60%
80%
100%
CBA
20%
40%
60%
80%
100%
CBA
White = 121-200 ms
Grey = >200 ms0%
0 min 20 min0%
0 min 20 min
RO FLR
CFE Stability – Verma et al, H t Rh th 8Heart Rhythm 2008
150
80
90100
*Range A, r=0.82
75
100
125
0 m
in
2030
4050
60
7080 *
%
5050 75 100 125 150
20 min
A B
010
A B C
Degree of Overlap 0/20 min Cycle Length 50-120 ms
160
200
0 m
in
400
500
600
700
800
0 m
in
Range C, r=0.08Range B, r=0.21
120120 160 200
20 min
200
300
400
200 300 400 500 600 700 800
20 min
C DCycle Length >200 msCycle Length 121-200 ms
Substrate vs Trigger ggAblation for Reducing Atrial FibrillationAtrial Fibrillation
A M l i R d i d T i lA Multicenter, Randomized Trial
Principal Investigator:p gAtul Verma, MD FRCPCSouthlake Regional Health Centre, Canada
Presented at Heart Rhythm 2009 – Late Breaking Trials
Study PurposeStudy Purpose
To compare the efficacy of three AF ablation To compare the efficacy of three AF ablation strategies:– Targeting the triggers of AF via PV isolation (PVI) – Targeting the substrate of AF maintenance via
elimination of complex fractionated electrograms (CFE)A h b id h f PVI CFE bl i– A hybrid approach of PVI + CFE ablation
High‐burden paroxysmal (65%) and persistent (35%) AF populationAF populationMulticenter, randomized trial
PVI StrategyPVI Strategy
Wide, circumferential PV Wide, circumferential PV antral isolationLesions placed >1‐2 cm poutside of the PV ostiaEndpoint of entrance pblock of all PV antra as documented by a
lcircular mapping catheter
CFE Strategy ‐ ICFE Strategy I
Spontaneous or induced Spontaneous or induced AF (must persist >1 min)Automated CFE mapping algorithmAll CFE regions targeted g g(CL<120 ms)Target all CFE regions in LA, then CS, then RA
CFE Strategy ‐ IICFE Strategy II
Endpoint of elimination of all pCFE sites or termination of AF and non‐inducibility of AFIf AF did t t i t ft If AF did not terminate after all CFE sites ablated, cardioversion allowedIf AF terminated to AT or AFL, this was mapped/ ablated when possibleablated when possibleIV antiarrhythmics not used during ablationg
PVI+CFE StrategyPVI+CFE Strategy
PVI completed firstPVI completed firstThen, CFE mapped (spontaneous or induced ( pAF) and ablatedEndpoint of PVI followed pby AF termination/ non‐inducibilityIf AF not terminated after all CFE ablated,
di i ll dcardioversion allowed
Procedural Details ‐ IIProcedural Details II
CFE PVI PVI+CFE pp
Procedure Time (min)
224 ± 80 181 ± 74 225 ± 68 NS( )
Mapping Time (min)
39 ± 18 29 ± 21 41 ± 20 0.09
Fluoroscopy Time (min)
56 ± 28 58 ± 27 60 ± 34 NS
RF Time (min)
65 ± 33 68 ± 41 77 ± 45 NS
Freedom from AF/AFL/AT(1 procedure)
AT
p=0.002
m A
F/A
FL/A
PVI+CFEFree
dom
fro
PVI+CFEPVICFE
F
5.6% on antiarrhythmics, evenly distributed between groups
Months Post-Ablation
Freedom from AF/AFL/AT(2 procedures)
100 p<0.00010 04
70
80
90
AT
%
p=0.04
50
60
70
om A
F/A
FL/A
PVI+CFEPVICFE
20
30
40
Free
dom
fro CFE
0
10
M1 M2 M3 M4 M5 M6 M7 M8 M9 M10 M11 M12M1 M2 M3 M4 M5 M6 M7 M8 M9 M10 M11 M12
Months post initial ablation
5.6% on antiarrhythmics, evenly distributed between groups
Repeat ProceduresRepeat Procedures100
70
80
90
dure
s p=0.008
4750
60
70
epea
t pro
ced
p=0.07
p=0.21
3130
40
nder
goin
g re
15
10
20% u
n
0PVI+CFE PVI CFE
Paroxysmal/Persistent bSubgroups
100 p=0.03 p=0.03
70
80
90
100
hmia
%
p=0.14 p=0.14
p p
40
50
60
70
rom
arr
hyth
PVI+CFEPVI CFE
10
20
30
40
Free
dom
fr CFE
0
10
AF AF/AFL/AT AF AF/AFL/AT
P l P i t tParoxysmal Persistent
ConclusionsConclusions
In high burden paroxysmal/persistent AF patients, In high burden paroxysmal/persistent AF patients, PVI+CFE is associated with the highest freedom from atrial arrhythmia at one yearPVI+CFE requires fewer repeat procedures compared to either strategy aloneCFE is associated with the highest recurrence rate and highest number of repeat proceduresBenefit of PVI+CFE may be more pronounced in persistent AF
Ganglionated Plexi (GPs)
Ablating Autonomic InputsAblating Autonomic Inputs
Evidence from Pappone et al that vagal denervation may be pp g yan important reason for AF cure by CARTO‐guided approach
Pappone et al,Circulation 2004
Targeting GP InputsTargeting GP Inputs
Recently, Sherlag et al, JICE 2005 reported improved Recently, Sherlag et al, JICE 2005 reported improved success rates with ablation of GPs in addition to PV isolation– 90% vs 71% success respectively
Sherlag et al, JACC 2005Sherlag et al, JACC 2005
GP input may beimportant increating acreating a substrate for converting PVconverting PVfiring into AF.
Correlation betweenCFE and regionsof autonomic inputof autonomic input.
Mixed Clinical ResultsMixed Clinical Results
Scanavacca et al, Circulation 2006Scanavacca et al, Circulation 2006– Vagal reflex‐guided GP ablation in paroxysmal AF (n=10)– 29% success rate after one procedurep
Pokushalov et al, Heart Rhythm 2009– Vagal reflex‐guided GP ablation vs anatomic ablation of
GP sites in paroxysmal AF (n=40)S l ti GP bl ti l %– Selective GP ablation success only 43%
– Anatomic GP ablation success 77%
Ganglionated PlexiGanglionated Plexi
Persistent AF population is the next big horizonPersistent AF population is the next big horizonWill not be adequately addressed by the “one size fits all” approachppNew mapping‐based targets need to be assessed
Dominant Frequency (DF)
Dominant FrequencyDominant Frequency
May represent specific rotor sites responsible for AF May represent specific rotor sites responsible for AF perpetuationMay be targeted as a lone or combined hybrid y g ystrategy
DF DistributionDF Distribution
PAROXYSMAL
Sanders et al Circulation 2005
CHRONIC
Sanders et al, Circulation 2005
Atienza et al, Heart Rhythm 2009Atienza et al, Heart Rhythm 2009
A t d ti iAcute reduction inDF in all chambers associated with higher freedom from AF long-term.
Ablation of DFmax sites associated
ith hi h f dwith higher freedom from AF (88% vs 30%)
Overall success rate 88% parox and 56% persistent
SummarySummary
EGM‐guided ablation is a promising avenue for g p gadjuvant ablation in addition to PVIMay be particularly useful in persistent AF
l b l h h b d lpopulation, but also higher‐burden paroxysmalsCFE offers the most promising target at present, but need for more refined definitionsneed for more refined definitionsGP ablation may overlap a lot with CFE and data still lackinggDF ablation may be promising, particularly in refining CFE sites